(M1330-01-05) Mixing Differences during Preparation Methods have Implications on Successful Formation of Amorphous Solid Dispersions

Purpose: The purpose of this study is to understand why different thermal and solvent-based methods can lead to different outcomes when trying to prepare same drug-polymer amorphous solid dispersion (ASDs). In both methods, ASDs are formed by first generating a homogeneous drug-polymer intermediate (Step 1 in Figure 1). Next, the conditions are changed to co-solidify the mixture (Step 2 in Figure 1). Differences in the phase behavior of the final product may be due to differences in the mixing dynamics during step 1 and/or the solidification kinetics during step 2. A collection of drug-polymer mixtures were previously observed to be not dispersible when melt-quenched but were dispersible when spray-dried. It was hypothesized that the original melt quenching process failed to yield a dispersion because a homogeneous mixture in the molten state was not formed during step 1, as opposed to failing because a homogeneous intermediate phase separated during cooling. To test this hypothesis, spray-dried ASDs which were determined to be homogeneous were re-processed using the melt-quenching process. Since these samples were already homogeneous and the cooling conditions were the same as the previous melt-quenched samples, the outcomes of this study will demonstrate whether step 1 or step 2 is responsible for the failure of the melt-quenching process for these systems.

Methods: Physical mixtures of three drugs (15% w/w): nifedipine (NIF), cimetidine (CIM), and terfenadine (TER), were prepared with polyvinylpyrrolidone vinyl acetate (PVPVA) co-polymer, and dissolved in methanol. A Yamato ADL311S with GAS 410 solvent recovery system was used to spray-dry the drug-PVPVA-methanol solutions. The inlet temperature was 90°C, blower speed was 0.64 m³/min, atomizing pressure was 0.1 MPa, and the feed rate was 12.5 mL/min. Spray-dried preparations were subsequently melt-quenched to confirm the effect of mixing. Spray-dried ASDs were added to a crucible and heated in a silicone oil bath at temperatures 10°C-15°C above the melting temperature of each drug (T_m,_drug). Once molten, the mixtures were stirred periodically using a glass rod. After 25-30 min, the crucible was placed in an ice water bath under nitrogen purge until the molten mixtures solidified (1-2 min). After preparation, all the samples were stored in a P<sub>2O₅ desiccator (~0% RH) until characterization.

Results: Dispersibility (defined as persistence of a single homogeneous amorphous phase, 1 day after preparation) of each drug in PVPVA was inferred from powder X-ray diffraction (PXRD), differential scanning calorimetry (DSC), and polarized light hot stage microscopy data. Co-solidified mixtures were considered dispersed if characterization by all three methods indicated non-crystalline and homogeneous systems. If one or more techniques revealed either crystallinity or phase separation, samples were classified as not dispersed. Results for the drug-polymer mixtures are provided in Table 1. Melt-quenched physical mixtures of 15% w/w NIF, CIM, and TER in PVPVA were not fully dispersed. When drug-polymer mixture samples were spray-dried prior to melt-quenching, however, these melt-quenched samples were characterized as dispersed in the PVPVA. Therefore, dispersibility for NIF-PVPVA, CIM-PVPVA, and TER-PVPVA melt-quenched samples changed from not dispersible to dispersible when spray-dried samples were used instead of physical mixtures. As the quenching process remained consistent between the original melt-quench studies and the melt-quenching studies using spray-dried dispersions, the change in the dispersibility outcome suggests that a homogenous mixture was not formed in the original melt-quench studies. This difference in the dispersibility categorization between the original melt-quenched sample and the original spray-dried sample could be because the solvent facilitated formation of a homogeneous drug-polymer mixture by enhancing the mixing kinetics.

Conclusion: The current results support the hypothesis that the original melt-quenching process fails because it does not produce a homogeneous dispersion in the molten state. These results are a significant step toward fully understanding the mechanisms by which processing conditions contribute to successful formation of amorphous dispersions. This study emphasizes that the success of a melt-quench process depends on effective mixing of the drug and polymer as well as the scale of mixing in the feed blend.

Figure 1: Two-step ASD preparation using fusion and solvent evaporation methods.

Table 1: Dispersibility outcomes of the drug-polymer combinations when prepared by different methods.